The present embodiments relate to medical treatment planning. The goal of radiation treatment planning is to calculate a set of parameters to control a radiation therapy device for delivering radiation to a patient. Ideally, all of the radiation delivered is concentrated within a tumor, and no radiation is delivered outside the tumor. Practically, this is not likely. Therefore, radiation treatment planning attempts to find a solution that trades off radiation dose to the tumor and radiation dose outside the tumor.
A human operator may set the radiation treatment parameters, such as number of beams, beam location, beam direction, and beam strength. The predicted radiation dosages to the tumor and outside the tumor are computed from the operator set radiation treatment parameters. Inverse treatment planning systems (ITPS) have been developed that attempt to create an optimal solution of radiation treatment parameters given an objective function. This objective function measures how well a particular radiation plan meets the goals of the treatment.
The objective function is based either on dose or on a generic biological model. Objective functions based on dose require an input as to what the desired dose should be within a defined tumor, and what the desired dose should be outside the tumor. The ITPS then tries to optimize a solution to meet these criteria. Objective functions based on biological parameters include a generic model for how different dose distributions affect a tumor and how different dose distributions affect normal tissue outside the tumor. The input to the system is the desired tumor control probability (TCP) and normal tissue complications probability (NTCP). The system tries to maximize TCP and minimize NTCP to meet the desired goal. This is done by using the biological model to extract the desired dosage to the tumor, and allowed tolerances of dose to normal tissue, to achieve these TCP and NTCP goals. Given the dosages, the radiation treatment parameters are determined. However, the radiation treatment parameters may not be optimal for a given patient.